Take two readings a minute apart, morning and evening, for about 7 days. Bring every number, the averages, your notes, and the cuff itself.
Bring a week of home readings taken the standard way: two readings one minute apart, every morning and every evening, for about seven days. Write down every number with the date and time, add the morning and evening averages, note anything unusual, and bring the cuff itself so it can be checked against the clinic's device.
From my side of the desk, a good log answers one question fast: what is this person's usual blood pressure at home?
The American Medical Association and Target:BP (a joint AHA and AMA program) describe the same routine for self-measured blood pressure, often shortened to SMBP:
| What | How |
|---|---|
| Readings per session | 2, one minute apart |
| Sessions per day | 2, morning and evening |
| Days | 7 in a row |
| Total | 28 readings |
The AMA calls seven days optimal. If your week gets cut short, a shorter run still helps, though the sources don't agree on the floor. Target:BP says two consecutive days is acceptable, while the National Association of Community Health Centers (NACHC) toolkit lists at least 3 days (12 readings) as the minimum. More days give a steadier average, so aim for seven and don't throw out a partial week.
The setup for each reading matters as much as the schedule. Sit quietly for a few minutes first, back supported, feet flat, arm resting at about heart level, and no talking. I've laid that out step by step in how to measure blood pressure at home.
Mostly, the average. Target:BP's instructions are to average all the systolic and diastolic readings for the week into one systolic number and one diastolic number, and the NACHC toolkit describes that overall average as the figure used for diagnosis and treatment decisions. A single 162 on Thursday matters much less than where 28 readings land together.
Some protocols drop it. A 2022 review of the "722" protocol (7 days, 2 sessions, 2 readings) in the Journal of Clinical Hypertension explains that European guidance discards the first day, because measurement bias is more likely at the start of a monitoring stretch. The same review describes the AHA/AMA version as 7 days, or 8 if the first day is discarded. Target:BP's own how-to page doesn't mention dropping it.
So whether day 1 counts depends on your doctor's protocol. Record it either way. Your doctor can leave out a day you wrote down, but they can't use one you skipped. If you want to know why first readings tend to run high, see why your first blood pressure reading is often higher.
Doctors often look at the morning and evening averages separately, because the two can differ and medication timing can affect either one. That's why the time of each reading belongs in the log. There's more on this in morning vs evening blood pressure.
Here's what I'd want to see on every line:
| Include | Why it helps |
|---|---|
| Date and exact time | Lets your doctor split morning from evening and line readings up with when you took medication |
| Both readings from each session | Shows how much your numbers move from one minute to the next |
| The exact numbers on the screen | Rounding 143/87 to "about 140 over 90" shifts the average |
| Which arm, and sitting or lying | Switching arms midweek adds noise you can't see later |
| Pulse, if your monitor shows it | Extra context your doctor may want |
| Notes | Missed dose, headache, poor sleep, pain, coffee, a stressful morning |
At the bottom, add three averages: morning, evening, and overall. If your doctor follows a drop-day-1 protocol, it's easy for them to recompute, but having your numbers there saves time in a short visit.
Target:BP also suggests keeping a written log in addition to the monitor's memory, as a backup. Monitor memory has no notes, and if two people share the cuff, it has both of you in it.
A printable blood pressure log works fine. So does a blood pressure app, as long as it keeps the time stamp and lets you add notes.
These are all common, and each one makes the average harder to trust.
Bring the monitor along with the log. Harvard Health suggests asking a nurse to take one reading with your monitor and one with the office device. The AMA's guidance is that home readings should come from a validated upper-arm device with a cuff sized to your arm, and the NACHC toolkit lists checking a home monitor against a more accurate machine as a routine care-team task.
That check answers two practical questions. Is your device reading in the right neighborhood, and is the cuff the right size for your arm? Expect the two devices to differ a little. Your doctor or nurse can tell you whether the gap is big enough to matter.
Some readings shouldn't sit in a log until your next visit. Cleveland Clinic's guidance:
If your averages are running well above the target your doctor gave you, call the office rather than waiting weeks for a scheduled visit.
Before the week starts
During the week
Before the visit
Bring it, but don't rely on it alone. Target:BP recommends a written log with device memory as the backup. The memory won't tell anyone about the missed pill or the 3 a.m. phone call.
Keep them in and add a note, such as "cuff slipped" or "talking on the phone." Deleting readings by hand is how a log ends up looking better than the person's blood pressure.
Ideally, finish the seven days in the week or two right before the visit, so the numbers reflect your current medications and routine.
It can be better, if it records the time of every reading, keeps both readings from each session, and lets you add notes. The format matters less than following the protocol consistently.
BP Log's guided flow times two readings a minute apart and averages them automatically, and you can tag a reading with things like a missed medication or caffeine. The free doctor visit report PDF puts your averages and an AM vs PM chart on page 1, then lists every reading, split AM and PM with averages, from page 2. You can also export everything as a CSV file.